American Accounting Association


Session Title: Linguistic Modeling and Feature Extraction in Fraudulent Reporting
Presentation Date: Wednesday August 10, 2011
Presentation Time: 10:15 am-11:45 am

The Exogenous Issue of Feature Extraction

Rua-Huan Tsaih, National Chengchi University MIS Department
Wan-Ying Lin, National Chengchi University Department of Accounting
Shin-Ying Huang, National Chengchi University MIS Department

ABSTRACT: This study applies Growing Hierarchical Self-Organizing Map (GHSOM) to fraudulent financial reporting (FFR) samples to explore the exogenous issue of feature extraction. Based upon certain input variables derived from the FFR literature and statistical tool, FFR samples are classified into several small-sized leaf nodes of GHSOM. For each GHSOM leaf node, this study uncovers common fraudulent techniques from corresponding FFR indictments and sentences (exogenous information) of clustered samples without referring to the attributes of input variables. As different leaf nodes have distinctive common fraudulent techniques, this study confirms that GHSOM can provide implicitly a relationship between common fraud techniques (an exogenous variable) and input variables. It is also demonstrates that GHSOM extracts features from exogenous information that are more abundant and more informative than input variables and classifies exogenous variables in terms of input variables.

Full-Text is no longer available online. Please contact the author(s) for more information about this manuscript.

Back to Session Listing